Explorative Data Analysis

Exploring AirBnB listings in Paris Source: https://www.kaggle.com/vaibhav2025/airbnb-listings-reviews Dataset file is original one preprocessed by PrepData notebook

Importing libraries, we'll start with numerical and text columns followed by categorical columns and maps

import pandas as pd
import numpy as np
import matplotlib
from matplotlib import pyplot as plt
import seaborn as sns
import plotly.express as px
import matplotlib as mp
from wordcloud import WordCloud,STOPWORDS
from PIL import Image
import folium
import folium.plugins as plugins

#reading the prepared dataset
df = pd.read_csv('Airbnb_Data/prepListings.csv', low_memory=False)
df

	listing_id	name	host_since	host_location	host_is_superhost	host_total_listings_count	host_identity_verified	neighbourhood	latitude	longitude	...	Ethernet connection	balcony	Private_entrance	TV	Dedicated_workspace	Refrigerator	Cooking	Bathtub	Shower	price_category
0	281420	Beautiful Flat in le Village Montmartre, Paris	2011-12-03	Paris, Ile-de-France, France	f	1.0	f	Buttes-Montmartre	48.88668	2.33343	...	False	False	False	False	False	False	False	False	False	Cheap
1	3705183	39 m² Paris (Sacre Cœur)	2013-11-29	Paris, Ile-de-France, France	f	1.0	t	Buttes-Montmartre	48.88617	2.34515	...	False	False	False	False	False	False	False	False	False	Average
2	4082273	Lovely apartment with Terrace, 60m2	2014-07-31	Paris, Ile-de-France, France	f	1.0	f	Elysee	48.88112	2.31712	...	False	False	False	True	False	False	False	False	False	Average
3	4797344	Cosy studio (close to Eiffel tower)	2013-12-17	Paris, Ile-de-France, France	f	1.0	t	Vaugirard	48.84571	2.30584	...	False	False	False	True	False	False	False	False	False	Cheap
4	4823489	Close to Eiffel Tower - Beautiful flat : 2 rooms	2014-12-14	Paris, Ile-de-France, France	f	1.0	f	Passy	48.85500	2.26979	...	False	False	False	True	False	False	False	True	False	Cheap
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
47463	38338635	Appartement T2 neuf près du tram T3a Porte Didot	2015-04-13	Paris, Ile-de-France, France	f	1.0	t	Observatoire	48.82701	2.31419	...	False	False	False	True	True	False	False	False	False	Average
47464	38538692	Cozy Studio in Montmartre	2013-11-27	Paris, Ile-de-France, France	f	1.0	t	Buttes-Montmartre	48.89309	2.33206	...	False	False	False	True	False	False	False	False	False	Cheap
47465	38683356	Nice and cosy mini-appartement in Paris	2012-04-27	Paris, Ile-de-France, France	f	1.0	t	Buttes-Montmartre	48.88699	2.34920	...	False	False	False	False	True	False	False	False	False	Cheap
47466	39659000	Charming apartment near Rue Saint Maur / Oberk...	2015-07-16	Paris, Ile-de-France, France	f	1.0	t	Popincourt	48.86687	2.38123	...	False	False	False	True	True	False	False	False	False	Average
47467	40219504	Cosy apartment with view on Canal St Martin	2013-06-17	Paris, Ile-de-France, France	f	1.0	t	Enclos-St-Laurent	48.87217	2.36320	...	False	False	False	False	True	True	True	False	True	Cheap

47468 rows × 39 columns

Density plot with dates of host's first listing

df['host_since'] = pd.to_datetime(df.host_since)
sns.set_style('whitegrid')
sns.kdeplot(np.array(df.host_since))

<AxesSubplot:ylabel='Density'>

Count of host's listings - density

fig = plt.figure(figsize=(16,14))
df['price'].plot(kind='kde')
plt.xlim(0,4000)
plt.xlabel('Price', fontsize = 20)
plt.ylabel('Density', fontsize = 18)
plt.savefig('priceDist.png', dpi=1500)

Wordcloud graph with most used words in the "name" column

house = np.array(Image.open('Airbnb_Data/domek.jpg'))
df['name'] = df.name.apply(str)
names = ' '.join(df['name'].str.lower().values)
wordcloud = WordCloud(background_color='white',
                        height=900,
                        width=800,
                        scale=20,
                        prefer_horizontal=0.9999,
                        mask=house
                     ).generate(names)
plt.figure(figsize=[25, 18])
plt.imshow(wordcloud)
plt.axis('off')
#plt.savefig('WordName.jpg', dpi=900, bbox_inches='tight', quality=95)
plt.show()

Number of listings in neighbourhoods

Top 15 neighbourhoods by number of listing

fig = plt.figure(figsize=(16,14))

ax = df.groupby('neighbourhood').nunique()['listing_id'].sort_values(ascending=False)[:15].plot(kind='bar',colormap='PRGn')
def barValuesVert(ax, spacing=5):

    for rect in ax.patches:
        y_value = rect.get_height()
        x_value = rect.get_x() + rect.get_width() / 2
        space = spacing
        va = 'bottom'


        label = "{:d}".format(y_value)
        ax.annotate(
            label,
            (x_value, y_value),
            xytext=(0, space),
            textcoords="offset points",
            ha='center',
            va=va)

barValuesVert(ax)
plt.ylabel('Listings count',fontsize=15)
plt.xlabel('Neighbourhood',fontsize=15)
plt.xticks(fontsize = 14)
plt.grid
plt.yticks(np.arange(500,5501,500))


plt.show()

AVG price by neighbourhoods

Top 10 most expansive neighbourhoods by average

#average price in neighbourhoods
fig = plt.figure(figsize=(16,14))

ax = df.groupby('neighbourhood').mean()['price'].sort_values(ascending=False)[:10].plot(kind='bar',color = 'lightgreen')
def barValuesVert(ax, spacing=5):

    for rect in ax.patches:
        y_value = rect.get_height()
        x_value = rect.get_x() + rect.get_width() / 2
        space = spacing
        va = 'bottom'


        label = "{:f}".format(y_value)
        ax.annotate(
            label,
            (x_value, y_value),
            xytext=(0, space),
            textcoords="offset points",
            ha='center',
            va=va)

plt.ylabel('Avg price')
plt.xlabel('Neighbourhood')
barValuesVert(ax)
#plt.yticks([1000,2000,3000,4000,5000,6000,7000])
plt.savefig('PriceCityParts.png', dpi=1500)
plt.show()

Price distribution by type of room

#price according to room types
g = sns.catplot(x='room_type', y='price', data=df, height=8)
g.set_axis_labels('Room type', 'Price')

<seaborn.axisgrid.FacetGrid at 0x1af0c4a37c0>

Map with listings - one with price separated values and one heatmap

from folium.plugins import HeatMap
mapPoint = folium.Map([48.86,2.38],zoom_start=12)
for index, row in df.iterrows():
    if df.price_category[index] == 'Cheap':
        folium.CircleMarker( location=[df.latitude[index], df.longitude[index]],color='blue', radius=1 ).add_to( mapPoint )
    elif df.price_category[index] == 'Average':
        folium.CircleMarker( location=[df.latitude[index], df.longitude[index]],color='orange', radius=1 ).add_to( mapPoint )
    elif df.price_category[index] == 'Expansive':
        folium.CircleMarker( location=[df.latitude[index], df.longitude[index]],color='red', radius=1 ).add_to( mapPoint )



#mapPoint.save( 'PointMap.html')
display(mapPoint)

Make this Notebook Trusted to load map: File -> Trust Notebook

df['room_type'].value_counts()

Entire place    40771
Private room     5559
Hotel room        853
Shared room       285
Name: room_type, dtype: int64

mapPoint = folium.Map([48.86,2.38],zoom_start=12)
for index, row in df.iterrows():
    if df.room_type[index] == 'Entire place':
        #folium.CircleMarker( location=[df.latitude[index], df.longitude[index]],color='blue', radius=1 ).add_to( mapPoint )
        continue
    elif df.room_type[index] == 'Private room':
        folium.CircleMarker( location=[df.latitude[index], df.longitude[index]],color='orange', radius=1 ).add_to( mapPoint )
    elif df.room_type[index] == 'Hotel room':
        folium.CircleMarker( location=[df.latitude[index], df.longitude[index]],color='red', radius=1 ).add_to( mapPoint )
    elif df.room_type[index] == 'Shared room':
        folium.CircleMarker( location=[df.latitude[index], df.longitude[index]],color='blue', radius=1 ).add_to( mapPoint )



#mapPoint.save( 'PointMap.html')
display(mapPoint)

Make this Notebook Trusted to load map: File -> Trust Notebook

mapHeat = folium.Map([48.86,2.38],zoom_start=12)
HeatMap(df[['latitude','longitude']].dropna(),radius=8,gradient={0.2:'blue',0.4:'purple',0.6:'orange',1.0:'red'}).add_to(mapHeat)

#mapHeat.save( 'HeatMap.html')
display(mapHeat)

Make this Notebook Trusted to load map: File -> Trust Notebook

Histograms of numerical features

fig = plt.figure(figsize = (15,10))
ax = fig.gca()
df.hist(ax=ax)
plt.show()

<ipython-input-34-945c0f86f323>:3: UserWarning: To output multiple subplots, the figure containing the passed axes is being cleared
  df.hist(ax=ax)